[论文解读] A New Central Limit Theorem for the Augmented IPW Estimator: Variance Inflation, Cross-Fit Covariance and Beyond

Kuanhao Jiang, Rajarshi Mukherjee+2 more|arXiv (Cornell University)|May 20, 2022

Advanced Causal Inference Techniques被引用 4

一句话总结

本文在高维设定下建立了交叉拟合增广逆概率加权（AIPW）估计量的新中心极限定理，无需假设稀疏性。它揭示了两个关键现象：显著的方差膨胀，其程度取决于信噪比；以及预交叉拟合估计量之间非可忽略的渐近协方差，这两者在经典渐近理论中均不存在。证明方法结合了近似消息传递、确定性等价物和留一法分析。

ABSTRACT

Estimation of the average treatment effect (ATE) is a central problem in causal inference. In recent times, inference for the ATE in the presence of high-dimensional covariates has been extensively studied. Among the diverse approaches that have been proposed, augmented inverse probability weighting (AIPW) with cross-fitting has emerged a popular choice in practice. In this work, we study this cross-fit AIPW estimator under well-specified outcome regression and propensity score models in a high-dimensional regime where the number of features and samples are both large and comparable. Under assumptions on the covariate distribution, we establish a new central limit theorem for the suitably scaled cross-fit AIPW that applies without any sparsity assumptions on the underlying high-dimensional parameters. Our CLT uncovers two crucial phenomena among others: (i) the AIPW exhibits a substantial variance inflation that can be precisely quantified in terms of the signal-to-noise ratio and other problem parameters, (ii) the asymptotic covariance between the pre-cross-fit estimators is non-negligible even on the root-n scale. These findings are strikingly different from their classical counterparts. On the technical front, our work utilizes a novel interplay between three distinct tools--approximate message passing theory, the theory of deterministic equivalents, and the leave-one-out approach. We believe our proof techniques should be useful for analyzing other two-stage estimators in this high-dimensional regime. Finally, we complement our theoretical results with simulations that demonstrate both the finite sample efficacy of our CLT and its robustness to our assumptions.

研究动机与目标

• 在特征数与样本数均较大且相近的高维设定下，建立交叉拟合AIPW估计量的新中心极限定理。
• 在结果回归与倾向得分模型设定正确但不假设高维参数稀疏性的前提下，刻画AIPW估计量的渐近行为。
• 揭示并量化两个新现象：方差膨胀以及预交叉拟合估计量之间的非可忽略渐近协方差。
• 构建一个结合近似消息传递、确定性等价物和留一法技术的证明框架，用于分析高维情形下的两阶段估计量。

提出的方法

• 在具有轻尾分布的独立同分布协变量的高维渐近框架下，推导交叉拟合AIPW估计量的新中心极限定理。
• 利用近似消息传递理论分析AIPW构建中使用的岭正则化回归估计量的高维行为。
• 应用确定性等价物理论，刻画涉及高维回归系数的二次型的极限行为。
• 采用留一法以解耦依赖性，并控制两阶段AIPW框架中的估计误差。
• 在一般矩条件假设下，建立AIPW估计量的渐近正态性，无需对高维参数施加稀疏性或结构假设。
• 推导出渐近方差与协方差结构的显式表达式，表明其依赖于信噪比和问题特定参数。

实验结果

研究问题

• RQ1 在无稀疏性假设的高维设定下，交叉拟合AIPW估计量的渐近分布如何表现？
• RQ2 高维估计对AIPW估计量的方差有何影响？是否可被量化？
• RQ3 为何AIPW中预交叉拟合估计量在√n尺度上表现出非可忽略的渐近协方差，与经典理论相悖？
• RQ4 能否为AIPW建立一个能捕捉这些高维现象的新中心极限定理？
• RQ5 分析高维、非稀疏情形下的两阶段估计量（如AIPW）需要哪些新颖的技术工具？

主要发现

• AIPW估计量表现出显著的方差膨胀，其程度可精确地用信噪比及其他问题参数量化，与经典行为相悖。
• 预交叉拟合估计量之间的渐近协方差在√n尺度上非可忽略，这一现象未被经典渐近理论所捕捉。
• AIPW估计量的极限分布以传统意义而言非正态，其分布依赖于干扰参数估计量的联合分布。
• 与经典情形相比，AIPW估计量的渐近方差被放大，其放大因子取决于岭正则化参数及设计矩阵的协方差结构。
• 理论框架成功捕捉了模拟中观察到的有限样本行为，对模型误设和分布假设具有鲁棒性。
• 结合近似消息传递、确定性等价物和留一法分析的证明技术，可推广至高维设定下的其他两阶段估计量。